Improving Face Recognition Confidence by Superimposing Training Images
The folder "Dataset" contains the training images produced by running the superimposing.py script for each of the 40 subjects and it contains one test image of each of the 40 subjects.
"1.test.pgm" - Refers to the test image of subject 1 (s1).
"compiled.1.pgm" - Refers to the superimposed image of subject 1 (s1).
The same syntax is followed for the each of the 40 subjects.
Simply run the Code 'face_recognizer_proposed.py' and it will display the confidence values.
The 'prop_output.txt' contains the confidence values (output) of this approach.